40 Mr. Victor Horsley [Feb. 3, 



with our south-coast examples. In the remarkable town of Troesmis, 

 founded in the third century, near Trajan's wall at the mouth of the 

 Danube, these U -shaped towers are particularly well seen. 



The arrangements within the walls. — These fortified towns, of 

 course, enclose an area with numerous buildings, but so far very little 

 systematic excavation has been made. I have found myself that the 

 arrangement depended very much on the condition of the outside 

 wall ; and I may here digress for a moment to take up the question of 

 apparent deficiencies in the outside wall, which have always excited 

 much controversy among antiquarians. Thus at Eutnpius, Lemanus, 

 Regulbium, Anderida, and Garriononum, one wall is wanting, but as 

 regards Regulbium, there is extant a seventeenth century map, which 

 I now show you, exhibiting the fact that the castrum in question was 

 a perfect quadrangle, and the missing north wall has simply been 

 washed away by the sea. At Rutupia ruins of the wall have been 

 found on the cliff. At Anderida tho missing part is obviously just 

 beneath the turf and has been proved to be there by excavation ; it 

 may have been thrown down by the undermining processes of the 

 mediaeval siege which the place underwent. At Garriononum also 

 the foundations have been traced. We come then to the single in- 

 stance of Lemanus. This castrum, the walls of which it is admitted 

 on all hands have been dislocated and ruined by a landslip, stands 

 on the slope of a hill, and at its lower border, i.e. the south side, 

 there is a slightly elevated piece of ground which was taken by 

 Mr. Roach Smith, who made the first excavation here, to be a portion 

 of the landslip. Thinking one clay that I could detect in the bank 

 some of the flat foundation cross stones of an ordinary wall, I insti- 

 tuted a systematic excavation of this portion of the circumference of 

 the camp, and found that so far from its being an elevation merely 

 due to landslip, it was really a remarkably interesting example of a 

 revetted harbour wall. It consisted of a mass of layers of boulders 

 embedded in concrete, on the top of which the ordinary wall was 

 built. At this particular point the concrete was carried backwards 

 over the area of the camp for forty feet, so as to form a large and 

 immovable platform. It seems to me that this was possibly the 

 platform of a battery for defending the ships moored by the wall, as 

 represented on one of Trajan's coins. 



To return to the question of the buildings within these castra ; 

 as is shown in Trajan's column, there were numorous houses covered 

 with pent-house roofs and the well-known roofing tiles with imbri- 

 cated edges, such as are seen to be extensively used in many parts of 

 this country, especially in East Anglia, and which we certainly owe 

 to the Romans, as our Belgic and Celtic forefathers roofed their huts 

 with straw and thatch, such as may be also seen in many parts of the 

 country, especially well exemplified again in East Anglia. 



There were, of course, a forum and pretorian building, the 

 residence of the Governor, and barracks for the troops, basilica for 

 the Court of Justice, and private houses, markets, baths, etc. All 





